Alienware 18 (end 2013) crashing

[smartali89]

Hello,

It's been very long since I've posted here so apology in advance if I am posting in the wrong place.

I have this Alienware 18 with the following Config:

Core i7-4930MX
32GB Ram, 780M SLI, Samsung EVO 1TB SSD & 1TB M2

At first the games started crashing (mostly in SLI config) and sudden shut downs started. Since it was out of warranty and I am in no reach of a good hardware repair shop I tested and found out that the primary GPU had gone dead so I switched the GPU secondary to primary and it solved the basic gaming issues.

Then the laptop started to crash randomly without any load on it and even when I left it idle it would just crash and won't start on its own. Turning off the Turbo-boost helped solve this issue.

Today I took out all the Rams and tested them one by one using memtest, it crashed again so I went to bios and turned off Turbo boost and then all the tests ran fine.

I tried to do benchmarking using Sis Sandra, wPrime and it would crash as well when Turbo boost was enabled.

Then I gave stress using PassMark's BurnInTest with and without Turbo boost. What I noticed was without Turbo boost CPU temperature wasn't going beyond 90 and with it it was crossing over 95 and then sudden crashes. I also replaced the thermal grease and cleaned the heat sink but the issue persist.

How can I confirm if its a CPU fault? and if it is should I replace it?

 

[GhostRyder]

Hello,

It's been very long since I've posted here so apology in advance if I am posting in the wrong place.

I have this Alienware 18 with the following Config:

Core i7-4930MX
32GB Ram, 780M SLI, Samsung EVO 1TB SSD & 1TB M2

At first the games started crashing (mostly in SLI config) and sudden shut downs started. Since it was out of warranty and I am in no reach of a good hardware repair shop I tested and found out that the primary GPU had gone dead so I switched the GPU secondary to primary and it solved the basic gaming issues.

Then the laptop started to crash randomly without any load on it and even when I left it idle it would just crash and won't start on its own. Turning off the Turbo-boost helped solve this issue.

Today I took out all the Rams and tested them one by one using memtest, it crashed again so I went to bios and turned off Turbo boost and then all the tests ran fine.

I tried to do benchmarking using Sis Sandra, wPrime and it would crash as well when Turbo boost was enabled.

Then I gave stress using PassMark's BurnInTest with and without Turbo boost. What I noticed was without Turbo boost CPU temperature wasn't going beyond 90 and with it it was crossing over 95 and then sudden crashes. I also replaced the thermal grease and cleaned the heat sink but the issue persist.

How can I confirm if its a CPU fault? and if it is should I replace it?

Well that's unfortunate for such a nice machine. What it looks like to me is the CPU with Turbo enabled is hitting its temp wall of 100c which is probably causing it to throttle and then crash if it can't keep it in check. If you cleaned out the system and replaced the thermal grease then likely you have a cooling issue going on either with the heatsink or the fan. Check and see if the fans spin up properly for starters and then see how hot the air being blown out of the fans are (Compare fans on each side and speeds).

After that see what you find out and go from there.

 

[smartali89]

Hello,

Thank you for reply. As you suggested I checked the fan speed and throw and it seemed fine however it high-temperatures the fan didn't go fast resulting in reaching 95+ very soon and crash. I searched and found out that several others are facing this issue and then I found out that fan speed can be modified using HWInfo. I changed the fan-speed to max and then ran the test again. It lasted a little longer but got shut again.

This is most definitely cooling issue but I am confused why wouldn't the temperature rise beyond 80 without Turbo but goes 95 in less than a few seconds in Turbo?

Should I look for aftermarket heatsink and cooler?

 

[GhostRyder]

Hello,

Thank you for reply. As you suggested I checked the fan speed and throw and it seemed fine however it high-temperatures the fan didn't go fast resulting in reaching 95+ very soon and crash. I searched and found out that several others are facing this issue and then I found out that fan speed can be modified using HWInfo. I changed the fan-speed to max and then ran the test again. It lasted a little longer but got shut again.

This is most definitely cooling issue but I am confused why wouldn't the temperature rise beyond 80 without Turbo but goes 95 in less than a few seconds in Turbo?

Should I look for aftermarket heatsink and cooler?

Because Turbo allows the machine to boost its processor clocks and voltage up to a certain point. While I find the situation odd the way its handling it, it seems to be that its getting hot once it boosts. It could also be a bad chip (or unstable) at the higher clocks though that is more rare and would not make sense if it lasts longer based on higher cooling.

So with Turbo enabled and fan speed max on both fans the temps still hit 95? I feel there maybe some form of separation happening or the heatsink is damaged in some way. If you have time (Or can) open it up and take some pictures on the insides and heatsink and check to make sure the screws and contacts are fine (Including thermal pads) as that could be causing an issue.

Also since you have a bad video card and your sure its dead, I recommend removing it (And the SLI bridge) to eliminate it as a possibility for causing issues with the system.

 

[FreedomEclipse]

Its a laptop. You wont be able to find an aftermarket heatsink and cooler. Only thing you can do is look for the exact same parts on ebay or amazon to replace the ones you might think are faulty etc etc

 

[smartali89]

Because Turbo allows the machine to boost its processor clocks and voltage up to a certain point. While I find the situation odd the way its handling it, it seems to be that its getting hot once it boosts. It could also be a bad chip (or unstable) at the higher clocks though that is more rare and would not make sense if it lasts longer based on higher cooling.

So with Turbo enabled and fan speed max on both fans the temps still hit 95? I feel there maybe some form of separation happening or the heatsink is damaged in some way. If you have time (Or can) open it up and take some pictures on the insides and heatsink and check to make sure the screws and contacts are fine (Including thermal pads) as that could be causing an issue.

Also since you have a bad video card and your sure its dead, I recommend removing it (And the SLI bridge) to eliminate it as a possibility for causing issues with the system.

Yes, its going to 95. I did check all screws, heat sink and thermal paste, everything seemed fine. I already removed the 2nd GPU as the system won't power up when its connected.

Should I risk buying a new processor or Just try getting another heatsink? Because this issue has only started recently. I bought this machine two years ago and never had the CPU-heating issue.

What are your suggestions?

 

[OneMoar]

I think the board is bad the fact that It won't boot with the 2d gpu tells me to look there

 

[Cristovam]

I have a similar system and a very similar problem.
Core i7-4930MX
32GB Ram and 780M SLI, only different HDs.

It looks like overheating, I think the fans start too late and when the temperature rises to 100C the computer shuts down to avoid damage. In mine the GPUs are fine and (apparently) are not overheating.
This is not the first time this happens with me, last time a BIOS update sorted this out, but I have no idea when Dell is going to release a new version. =( I'll post if I find a way to fix this mess.

 

[Cristovam]

So, after using HWiNFO64 and manually configuring the fans speed, it passed the stress test.